curl-library
Re: Access violation in dprintf_formatf (on 100% progress?)
Date: Sun, 27 Sep 2015 17:55:12 +0200 (CEST)
On Sun, 27 Sep 2015, Thorsten Schöning wrote:
> This seems to be the right track, I didn't recognized it before but the
> progress output is completely broken if I use the curl versions of the
> *printf functions:
> 100 (nil) 1636520 (nil) 0 13970 (nil) (nil) 0 89551 0 --:--:--
The question is then why they are broken for you but work for everyone
else...?
> mprintf.h lets me choose if I want default functions to be replaced,
> curl_printf.h doesn't. Using "#error test" in mprintf.h I made sure that
> _MPRINTF_REPLACE is not defined.
>
> What's the reason behind this replacement and why is curl_printf doing it
> unconditionally?
Because it solves the problem for us internally when different printf()
implementations behave differently and since some implementations aren't very
good anyway we decided we need an implementation for them and thus we made it
easier for us to make all platforms use that code.
-- / daniel.haxx.se
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette: http://curl.haxx.se/mail/etiquette.html
Received on 2015-09-27